This project responds to RFA MH 13-260 Harnessing Advanced Health Technologies to Drive Mental Health Improvement, and is led by a New/Early Stage Investigator. Bipolar disorder (BD) and Schizophrenia (SZ) are leading causes of disability and are the costliest disorders to treat among serious mental illnesses. Ample evidence exists that evidence-based psychotherapies such as cognitive behavioral therapy (CBT) produce clinically significant improvements in symptoms of BD and SZ, and yet only 5% of patients in the community can access these treatments. Access to CBT for serious mental illnesses is limited by the restricted pool of trained providers, the resource intensity of these treatments, and cessation of reimbursement for psychological services in many public mental health systems. Our research in BD and SZ over the past five years indicates that mobile phones can provide automated yet personalized cognitive behavioral intervention that is feasible, acceptable, and associated with improvements in symptoms, medication adherence and socialization. Our intervention, called CBT2go, integrates ecological momentary assessment with personalized interventions delivered in the moment that symptoms and related experiences occur. We have developed an innovative platform that functions on various screen types, operating system, or data access technologies, and we have begun to incorporate innovative features into CBT2go, including location-based data and personalized predictive modeling, that can potentially lead to transformative mobile interventions. We propose a research study with the overarching aims of evaluating the effectiveness of CBT2go, gathering necessary data to refine its innovative features, and examining its costs, facilitators, and barries to implementation in a public mental health system. We propose a randomized controlled trial with three arms: 1) CBT2go, 2) an ecological momentary assessment only condition to control for self-monitoring and device contact, and 3) standard care. A total of 255 participants with either BD or SZ will be recruited from a large public mental health system that has minimal access to CBT. Participants will be assessed at baseline, 6 weeks (mid-treatment), 12 weeks (post-treatment) and 24 weeks (follow-up). The primary outcome of the study will be clinician rated global psychopathology, and secondary outcomes will include medication adherence, social functioning, and mental health service utilization. Recognizing that not all participants wil benefit, we propose to examine differential effectiveness in sub-groups. We will also examine mechanisms of change to inform mobile intervention design, focusing on the impact of CBT2go on cognitive insight and dysfunctional attitudes, which are two mediators of change in traditionally delivered CBT. Drawing from experience in implementation research, we will employ a mixed methods approach to assessing the facilitators, barriers, and costs of adoption of CBT2go from perspectives of consumers, technology experts, administrators and front-line clinicians. This study will lay the groundwork for CBT2go to be adapted and implemented with high fidelity in healthcare settings.